Hi list -
Can anybody tell me whether btrfs-raid10 reads and writes a stripe at a
time, like traditional raid10, or whether it reads and writes individual
redundant blocks like btrfs-raid1, but just locks particular disks as
mirror pairs, and/or locks the order in which mirror pairs should be
written to in sequence?
It occurs to me that this might be an important distinction for
workloads with very high small random I/O - stripe reads and writes
being at a disadvantage to individual blocks being read and written,
when the requests are smaller than the stripe.
Apologies again if this is a question that should have an obvious
answer, but as long as it took me to figure out that btrfs-raid1 is NOT
much like traditional raid1, it seemed worth asking.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html